AI Referral Traffic
AI referral traffic is website traffic that arrives when someone clicks a link inside an AI tool, such as ChatGPT, Perplexity, Gemini, Microsoft Copilot or Claude. It is usually a small share of total visits for most sites, but it is worth measuring separately because it shows whether AI assistants are citing and recommending your pages.
Where AI referral traffic comes from
AI tools that search the web often show source links alongside their answers. When a user clicks one, they land on your site. Common sources include:
- ChatGPT, which usually appears as
chatgpt.comand often addsutm_source=chatgpt.comto links. - Perplexity, which appears as
perplexity.ai. - Gemini, which appears as
gemini.google.com. - Microsoft Copilot, which appears as
copilot.microsoft.com. - Claude, which appears as
claude.ai.
Clicks from Google’s AI Overviews and AI Mode are different. They come through Google Search, so they are counted as normal Google organic traffic rather than as a separate referral.
How to track it in GA4
- In Google Analytics 4, open the Traffic acquisition report and set the dimension to Session source / medium.
- Filter for known AI domains to see how many sessions each sends.
- For a permanent view, create a custom channel group with an AI channel, using a regular expression that matches the AI source domains. Place it above the Referral channel so these sessions are not grouped with other referrals.
- Add landing page as a secondary dimension to see which pages AI tools send people to.
- Review the list of domains regularly, as new AI tools appear and existing ones change their domains.
Why some AI traffic goes missing
Not every visit from an AI tool is recorded as a referral. Mobile apps and desktop apps may not pass referrer information, so those visits can appear as direct traffic. Users may also copy a URL from an answer and paste it into a browser, or search for your brand after reading about it, which shows up as direct or branded search traffic. Your measured AI referral traffic is therefore likely to understate AI influence.
How to interpret it
- Look at engagement and conversions, not just sessions. Visitors from AI tools often arrive with a specific question already partly answered.
- Compare landing pages with the prompts you track to see which content AI tools cite.
- Combine referral data with prompt tracking and AI visibility monitoring for a fuller picture.
- Use attribution reports to see whether AI visits contribute to later conversions.
Common mistakes
- Leaving AI sources mixed in with general referral traffic.
- Assuming AI Overviews traffic will appear as a separate source in analytics.
- Treating low referral numbers as proof that AI tools have no influence.
- Comparing AI referral volumes directly with organic search without considering the much smaller user base for some tools.
Key Takeaways
- AI referral traffic comes from clicks on links inside AI tools such as ChatGPT and Perplexity.
- Clicks from Google AI Overviews and AI Mode count as Google organic traffic.
- A custom channel group in GA4 separates AI sources from other referrals.
- App visits and copied links often appear as direct traffic, so AI influence is understated.
- Look at landing pages, engagement and conversions, not just session counts.
Related Terms
Want to know where your site stands? Book a free SEO health check, or keep learning in the SEO Encyclopedia.